Loading video...

Video Failed to Load

There was a problem loading this video. This could be due to a temporary network issue or the video might be unavailable.

Transformer by hand ✍️ Given input X, this exercise goes through the key equations behind the 𝗦𝗲𝗹𝗳 𝗔𝘁𝘁𝗲𝗻𝘁𝗶𝗼𝗻 layer and the 𝗙𝗲𝗲𝗱 𝗙𝗼𝗿𝘄𝗮𝗿𝗱 layer, showing you how X is transformed into Y.

Tom Yeh

45,567 subscribers

15,488 views • 1 year ago •via X (Twitter)

Anya Rossi• Live Now

Private livecam show

3 Comments

EcoAstronomer1 year ago

Shouldn't V = Wv x chi?

Tom Yeh1 year ago

You are right! Thanks for spotting the error. 😅 You are really paying attention. 👍

SFTP.Callum1 year ago

Beautiful to see love laid out as a transformer

Related Videos

[Transformer] by Hand✍️📺 5-minute Video Tutorial Anna Rahn made this short video to explain the Transformer exercise for my Computer Vision course last spring. In 5 minutes, she demonstrates the key calculations of the Transformer by hand with pen and paper! Anna is a fantastic student. I am lucky to have her in my lab!

[Transformer] by Hand✍️📺 5-minute Video Tutorial Anna Rahn made this short video to explain the Transformer exercise for my Computer Vision course last spring. In 5 minutes, she demonstrates the key calculations of the Transformer by hand with pen and paper! Anna is a fantastic student. I am lucky to have her in my lab!

Tom Yeh

133,460 views • 2 years ago

MCP by hand ✍️ This exercise walks you through the key elements of the 𝗠𝗼𝗱𝗲𝗹 𝗖𝗼𝗻𝘁𝗲𝘅𝘁 𝗣𝗿𝗼𝘁𝗼𝗰𝗼𝗹 (MCP). Learn more at our workshop. Link below. 👇

MCP by hand ✍️ This exercise walks you through the key elements of the 𝗠𝗼𝗱𝗲𝗹 𝗖𝗼𝗻𝘁𝗲𝘅𝘁 𝗣𝗿𝗼𝘁𝗼𝗰𝗼𝗹 (MCP). Learn more at our workshop. Link below. 👇

Tom Yeh

33,683 views • 1 year ago

QuickSwap is now live on X Layer Mainnet ⚡️ The birth of a new Citadel on the dragon DEX, expanding the Polygon CDK family. Swap and LP on X Layer (farms coming soon) via QuickSwap to experience DeFi on the network at its very best. X Layer is ZK-powered Layer 2 network designed to connect the OKX and Ethereum communities, built using the Polygon CDK tech stack. It's designed to have seamless interoperability with other Layer 2 chains. Now plugged into the AggLayer, OKX’s massive community of 50+ million users is fully interoperable with all interconnected chains and can now tap into its unified liquidity. The network also uses zk-SNARK to make transactions significantly cheaper and far more compact. X Layer brings other benefits to users, such as near-instant finality, unified liquidity, robust security, high scalability... the list goes on. With this mainnet deployment and more to come, QuickSwap is leading the charge to ignite activity and growth on X Layer at rapid speeds. Going full throttle to further unite the Polygon | POL and OKX communities. This is only the beginning of a bright future for QuickSwap, X Layer, and the broader Polygon and Ethereum ecosystems. Get ready to rumble.

QuickSwap is now live on X Layer Mainnet ⚡️ The birth of a new Citadel on the dragon DEX, expanding the Polygon CDK family. Swap and LP on X Layer (farms coming soon) via QuickSwap to experience DeFi on the network at its very best. X Layer is ZK-powered Layer 2 network designed to connect the OKX and Ethereum communities, built using the Polygon CDK tech stack. It's designed to have seamless interoperability with other Layer 2 chains. Now plugged into the AggLayer, OKX’s massive community of 50+ million users is fully interoperable with all interconnected chains and can now tap into its unified liquidity. The network also uses zk-SNARK to make transactions significantly cheaper and far more compact. X Layer brings other benefits to users, such as near-instant finality, unified liquidity, robust security, high scalability... the list goes on. With this mainnet deployment and more to come, QuickSwap is leading the charge to ignite activity and growth on X Layer at rapid speeds. Going full throttle to further unite the Polygon | POL and OKX communities. This is only the beginning of a bright future for QuickSwap, X Layer, and the broader Polygon and Ethereum ecosystems. Get ready to rumble.

QuickSwap 🐲 DragonFi 2.0

20,857 views • 2 years ago

The sweat, effort, and emotion behind this poster were worth every second and you guys took it to another level! ♥️ Here’s a sneak peek into the layer by layer breakdown of one of my favourite posters from the #OG campaign!!!

The sweat, effort, and emotion behind this poster were worth every second and you guys took it to another level! ♥️ Here’s a sneak peek into the layer by layer breakdown of one of my favourite posters from the #OG campaign!!!

Suryahhh

21,244 views • 8 months ago

llm.c by Hand✍️ C programming + matrix multiplication by hand This combination is perhaps as low as we can get to explain how the Transformer works. Special thanks to Andrej Karpathy for encouraging early feedback and tetsuo //: 👾 for helping me understand the pragma magic. I hope this exercise can help people peak further into the LLM black box.

llm.c by Hand✍️ C programming + matrix multiplication by hand This combination is perhaps as low as we can get to explain how the Transformer works. Special thanks to Andrej Karpathy for encouraging early feedback and tetsuo //: 👾 for helping me understand the pragma magic. I hope this exercise can help people peak further into the LLM black box.

Tom Yeh

302,688 views • 2 years ago

This is how a freehand is born - layer by layer, painted by the artist’s hand, with no decals and no shortcuts. Want to add a freehand to your army, vehicle, or model? Message us: service@denofimagination.com #Freehand #Warhammer30k #HorusHeresy #Miniature #WarhammerCommunity

This is how a freehand is born - layer by layer, painted by the artist’s hand, with no decals and no shortcuts. Want to add a freehand to your army, vehicle, or model? Message us: [email protected] #Freehand #Warhammer30k #HorusHeresy #Miniature #WarhammerCommunity

Den of Imagination

12,230 views • 1 month ago

I gave the keynote at UC Berkeley's ProductCon - "How to Become an Elite AI-Powered Product Manager" This is the full talk + behind-the-scenes footage: 0:00 - Why This Matters 4:11 - Layer 1: Prompting Fundamentals 10:46 - Layer 2: AI Co-pilots 19:45 - Layer 3: AI Agents 27:00 - Layer 4: AI Prototyping 31:45 - Layer 5: AI Discovery 39:37 - Layer 6: Evals & Observability 42:53 - Layer 7: Vibe Experimentation 50:56 - Behind the Scenes

I gave the keynote at UC Berkeley's ProductCon - "How to Become an Elite AI-Powered Product Manager" This is the full talk + behind-the-scenes footage: 0:00 - Why This Matters 4:11 - Layer 1: Prompting Fundamentals 10:46 - Layer 2: AI Co-pilots 19:45 - Layer 3: AI Agents 27:00 - Layer 4: AI Prototyping 31:45 - Layer 5: AI Discovery 39:37 - Layer 6: Evals & Observability 42:53 - Layer 7: Vibe Experimentation 50:56 - Behind the Scenes

Aakash Gupta

410,662 views • 5 months ago

hyperliquids data layer allows you to see everyones positions look at the power of these api calls and this is how you can get a free api key

hyperliquids data layer allows you to see everyones positions look at the power of these api calls and this is how you can get a free api key

Moon Dev

16,476 views • 6 months ago

A “few” holes punched into a couple of problem greens today showing just how much water is trapped within the rootzone layer and zeolight added to improve infiltration rates through periods of high rainfall.

A “few” holes punched into a couple of problem greens today showing just how much water is trapped within the rootzone layer and zeolight added to improve infiltration rates through periods of high rainfall.

Prestbury GC Greens

50,410 views • 2 years ago

Have you ever seen the decapsulating of a microchip layer by layer using a 30W JPT Fiber Laser? Well now you have and the the precision is wild to see.

Have you ever seen the decapsulating of a microchip layer by layer using a 30W JPT Fiber Laser? Well now you have and the the precision is wild to see.

Bluntly Put Philosopher (BPP)

83,778 views • 3 months ago

Anna Rahn's video demonstration of Multi-Layer Perceptron (MLP) by hand ✍️ Anna made this short video for my computer vision course in the spring semester. Together with the spreadsheet exercise I posted yesterday, I hope they help more people understand deep neural networks!

Anna Rahn's video demonstration of Multi-Layer Perceptron (MLP) by hand ✍️ Anna made this short video for my computer vision course in the spring semester. Together with the spreadsheet exercise I posted yesterday, I hope they help more people understand deep neural networks!

Tom Yeh

120,435 views • 1 year ago

OKX Integrates Chainlink To Unlock $80 Trillion RWA Opportunity On X Layer OKX officially adopts Chainlink to anchor the Real-World Asset (RWA) infrastructure on its X Layer network. The integration provides the high-speed data feeds and cross-chain interoperability necessary to capture an estimated $80T in tokenized global markets. As a "Chainlink Scale" member, X Layer is now positioned as a primary destination for sovereign institutional-grade asset migration. The collaboration enables developers to build advanced decentralized applications that bridge legacy financial value with the "AI Agent Economy" on X Layer.

OKX Integrates Chainlink To Unlock $80 Trillion RWA Opportunity On X Layer OKX officially adopts Chainlink to anchor the Real-World Asset (RWA) infrastructure on its X Layer network. The integration provides the high-speed data feeds and cross-chain interoperability necessary to capture an estimated $80T in tokenized global markets. As a "Chainlink Scale" member, X Layer is now positioned as a primary destination for sovereign institutional-grade asset migration. The collaboration enables developers to build advanced decentralized applications that bridge legacy financial value with the "AI Agent Economy" on X Layer.

BSCN

16,358 views • 28 days ago

3/ How do you experience the GRID? The GRID unifies global intelligence into one layer, accessible through Sentient Chat. Every query to the GRID is addressed through complex workflows that draw on multiple agents, search, data sources, and tools within the network.

3/ How do you experience the GRID? The GRID unifies global intelligence into one layer, accessible through Sentient Chat. Every query to the GRID is addressed through complex workflows that draw on multiple agents, search, data sources, and tools within the network.

Sentient

157,847 views • 11 months ago

No, this isn’t a Transformer. It’s a Leopard 2 Leguan, folding its 26-meter bridge. This bridge layer is operated by engineers of the Royal Netherlands Army.

No, this isn’t a Transformer. It’s a Leopard 2 Leguan, folding its 26-meter bridge. This bridge layer is operated by engineers of the Royal Netherlands Army.

Klaas Meijer

48,417 views • 1 year ago

How to launch the Uniswap Labs 🦄 dApp from our web3 wallet and use X Layer for swaps at L2 prices

How to launch the Uniswap Labs 🦄 dApp from our web3 wallet and use X Layer for swaps at L2 prices

Haider

45,175 views • 6 months ago

The value is in the application layer. Success requires a people-first approach. Winston Weinberg reveals how Harvey built trust through deep industry knowledge and personalized solutions to become the first breakout application layer company in AI.

The value is in the application layer. Success requires a people-first approach. Winston Weinberg reveals how Harvey built trust through deep industry knowledge and personalized solutions to become the first breakout application layer company in AI.

Pat Grady

21,339 views • 1 year ago

A new era for the World Mobile ecosystem is loading. Network Builder brings decentralized coordination to the real world, and the first onchain auction goes live this December. This is ownership at the network layer. This is how you scale a movement. Stay ready, Mobilers.

A new era for the World Mobile ecosystem is loading. Network Builder brings decentralized coordination to the real world, and the first onchain auction goes live this December. This is ownership at the network layer. This is how you scale a movement. Stay ready, Mobilers.

World Mobile

100,192 views • 7 months ago

Soundness Layer is launching SOON! 🐬✨ Verifiable games, zk proofs, and a new era of onchain apps, all powered by Soundness layer, using Sui as the coordination layer and Walrus 🦭/acc as DDA. How ready are you to play in zero-knowledge? 🎮 #soundness

Soundness Layer is launching SOON! 🐬✨ Verifiable games, zk proofs, and a new era of onchain apps, all powered by Soundness layer, using Sui as the coordination layer and Walrus 🦭/acc as DDA. How ready are you to play in zero-knowledge? 🎮 #soundness

Soundness

38,677 views • 11 months ago

Look at how the thick paste gets transformed from blobs into a smooth, even layer in minutes thanks to the roller pressure. This is a manual coating or calendering process on a large industrial roller machine. Satisfying process to watch.

Look at how the thick paste gets transformed from blobs into a smooth, even layer in minutes thanks to the roller pressure. This is a manual coating or calendering process on a large industrial roller machine. Satisfying process to watch.

Johnny

16,310 views • 2 months ago